Skill++ keeps agent skill routers accurate as libraries scale
Adding a new skill can silently break routing for older skills because routers match on name and description, and similar metadata lets new skills crowd out existing ones. Builders maintaining growing agent skill libraries can apply the technique immediately: when adding a skill, describe how it differs from the closest skill already present. The approach requires no training data, making it practical for any router-based agent setup.
A paper on scaling agent skills, Skill++, rewrites only the name and description of each new skill added to an agent's skill library. The rewritten description states what the skill supports, how it differs from the closest existing skill, and what it excludes, with no router training and no labeled data. As a library grows to 731 skills over ten sessions, the router selects the required skill for 76.3% of queries versus 36.4% with original metadata. Substitution errors nearly disappear and crowding errors fall by half, with gains holding across three benchmarks, five routers, and the Codex CLI harness. The paper is available at arxiv.org/abs/2611.00544.
Good research on maintaining and scaling agent skills. DAIR.AI @dair_ai Great paper on scaling agent skills. It includes some great tips for keeping your agent skill library updated and running smoothly. (bookmark it) You will often see that adding a new skill to an agent can break routing for older skills, even when nothing about the older skills or the router changed. Routers pick skills by name and description. When a new skill has similar metadata, it can replace or crowd out an earlier one. Skill++ rewrites only the name and description of each incoming skill. The new description says what the skill supports, how it differs from its closest existing skill, and what it excludes. There is no router training and no labeled data. As a library grows to 731 skills over ten sessions, the router selects every required skill for 76.3% of queries, against 36.4% with the original metadata. Substitution errors nearly disappear, and crowding errors fall by half. The gains hold across three benchmarks and five routers, and they carry over to the Codex CLI harness. When you add a skill, describe how it differs from the closest skill you already have. Paper: arxiv.org/abs/2611.00544 Chat with Paper: academy.dair.ai/papers/skill-t… 🔗 View Quoted Tweet 💬 8 🔄 1 ❤️ 22 👀 3506 📊 11 ⚡
- DAIR.AI11-03 13:37原文
